Predicting accidental drug overdose as the cause of fatality in near real-time using the Suspected Potential Overdose Tracker (SPOT): public health implications
Bibliographic record
Abstract
Abstract Background Effective responses to the worsening drug overdose epidemic require accurate and timely drug overdose surveillance data. The objectives of this paper are to describe the development, functionality, and accuracy of the Suspected Potential Overdose Tracker (SPOT) for predicting accidental drug overdose as the cause and manner of death in near real-time, and public health implications of adopting the tool. Methods SPOT was developed to rapidly identify overdose deaths through a simple and duplicable process using data collected by death investigators. The tool assigns each death a ranking of 1 through 3 based on the likelihood of it being an unintentional drug overdose, with 1 representing the highest likelihood that the death will be confirmed as an unintentional drug overdose and 3 representing the lowest. We measured the accuracy of the tool for predicting overdose deaths by comparing potential overdose deaths in New York City from 2018–2020 that were identified using SPOT to finalized death certificates. We also calculated the proportion of death certificate-confirmed overdoses that were missed by the SPOT tool and the proportion of type 1 errors. Results SPOT captured up to 77% of unintentional drug overdose deaths using data collected within 72 h of fatality. The tool predicted unintentional drug overdose from 2018 to 2020 with 93–97% accuracy for cases assigned a ranking of 1, 87–91% accuracy for cases assigned a ranking of 2, and 62–73% accuracy for cases assigned a ranking of 3. Among all unintentional overdose deaths in 2018, 2019, and 2020, 21%, 28%, and 33% were missed by the SPOT tool, respectively. During this timeframe, the proportion of type 1 errors ranged from 15%-23%. Conclusions SPOT may be used by health departments, epidemiologists, public health programs, and others to monitor overdose fatalities before death certificate data becomes available. Improved monitoring of overdose fatalities allows for rapid data-driven decision making, identification of gaps in public health and public safety overdose response, and evaluation and response to overdose prevention interventions, programs, and policies.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.007 | 0.021 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.002 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.002 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.002 | 0.001 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".